SEARCH
NEW RPMS
DIRECTORIES
ABOUT
FAQ
VARIOUS
BLOG
DONATE


YUM REPOSITORY

 
 

ATOP

Section: User Commands (1)
Updated: January 2019
Index 

NAME

atop - Advanced System & Process Monitor 

SYNOPSIS

Interactive Usage:atop[-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y] [-C|-M|-D|-N|-A] [-afFG1xR] [-L linelen] [-Plabel[,label]...][interval[samples]]Writing and reading raw logfiles:atop-wrawfile[-a] [-S][interval[samples]]
atop-r [rawfile] [-b hh:mm] [-ehh:mm] [-g|-m|-d|-n|-u|-p|-s|-c|-v|-o|-y] [-C|-M|-D|-N|-A] [-fFG1xR] [-L linelen] [-Plabel[,label]...] 

DESCRIPTION

The programatopis an interactive monitor to view the load on a Linux system.It shows the occupation of the most critical hardware resources (from a performance point of view) on system level, i.e. cpu, memory, diskand network.
It also shows which processes are responsible for the indicatedload with respect to cpu and memory load on process level.Disk load is shown per process if "storage accounting" is active in the kernel.Network load is shown per process if the kernel module `netatop'has been installed.

Everyinterval(default: 10 seconds) information is shown about the resource occupationon system level (cpu, memory, disks and network layers), followedby a list of processes which have been active during the last interval(note that all processes that were unchanged during the last intervalare not shown, unless the key 'a' has been pressed or unless sorting onmemory occupation is done).If the list of active processes does not entirely fit onthe screen, only the top of the list is shown (sorted in order of activity).
The intervals are repeated till the number ofsamples(specified as command argument) is reached, or till the key 'q' is pressedin interactive mode.

When atopis started, it checks whether the standard output channel is connected to ascreen, or to a file/pipe. In the first case it produces screen control codes (via the ncurses library) and behaves interactively; in the second caseit produces flat ASCII-output.

In interactive mode, the output of atopscales dynamically to the current dimensions of the screen/window.
If the window is resized horizontally, columns will be added or removedautomatically. For this purpose, every column has a particular weight. Thecolumns with the highest weights that fit within the current width willbe shown.
If the window is resized vertically, lines of the process/thread list will be added or removed automatically.

Furthermore in interactive mode the output of atopcan be controlled by pressing particular keys.However it is also possible to specify such key asflagon the command line. In that caseatopswitches to the indicated mode on beforehand; this mode can be modified again interactively. Specifying such key as flag is especiallyuseful when runningatopwith output to a pipe or file (non-interactively).These flags are the same as the keys that can be pressed in interactivemode (see section INTERACTIVE COMMANDS).
Additional flags are available to support storage of atop-data in raw format (see section RAW DATA STORAGE). 

PROCESS ACCOUNTING

With every interval,atopreads the kernel administration to obtain information about allrunning processes.However, it is likely that during the interval also processes have terminated.These processes might have consumed system resources duringthis interval as well before they terminated.Therefor,atoptries to read the process accounting records that contain the accountinginformation of terminated processes and report these processes too.Only when the process accounting mechanism in the kernel is activated,the kernel writes such process accounting record to a filefor every process that terminates.

There are various ways foratopto get access to the process accounting records (tried in this order):

1.
When the environment variable ATOPACCT is set,it specifies the name of the process accounting file.In that case, process accounting for this fileshould have been activated on beforehand.Before opening this file for reading,atopdrops its root privileges (if any).
When this environment variable is present but itscontents is empty, process accounting will not be used at all.

2.
This is the preferred way of handling process accounting records!
When theatopacctddaemon is active, it has activated the process accounting mechanism inthe kernel and transfers to original accounting records to shadow files.In that case,atopdrops its root privileges and opens the current shadow file for reading.
This way is preferred, because theatopacctddaemon maintains full control of the sizes of the original processaccounting file (written by the kernel) and the shadow files (read by theatopprocesses). For further information, refer to theatopacctdman page.

3.
When theatopacctddaemon is not active, atopverifies if the process accounting mechanism has been switched onvia the separatepsacctpackage. In that case, the file/var/account/pacctis in use as process accounting file and atopopens this file for reading.

4.
As a last possibility,atopitself tries to activate the process accounting mechanism (requires rootprivileges) using the file/var/cache/atop.d/atop.acct(to be written by the kernel, to be read byatopitself). Process accounting remains active as long asat least oneatopprocess is alive.Whenever the lastatopprocess stops (either by pressing `q' or by `kill -15'), it deactivates theprocess accounting mechanism again. Therefor you should never terminateatopby `kill -9', because then it has no chance to stop process accounting.As a result, the accounting file may consume a lot of disk space after a while.
To avoid that the process accounting file consumes too much disk space,atopverifies at the end of every sample if the size of the process accountingfile exceeds 200 MiB and if thisatopprocess is the only one that is currently using the file.In that case the file is truncated to a size of zero.

Notice that root-privileges are required to switch on/off process accountingin the kernel. You can startatopas a root user or specify setuid-root privileges to the executable file.In the latter case,atopswitches on process accounting and drops the root-privileges again.
Ifatopdoes not run with root-privileges, it does not show informationabout finished processes.It indicates this situation with the messagemessage `no procacct` in the top-right corner (instead of the counter thatshows the number of exited processes).

When during one interval a lot of processes have finished,atopmight grow tremendously in memory when reading all process accountingrecords at the end of the interval. To avoid such excessive growth,atopwill never read more than 50 MiB with process information from theprocess accounting file per interval (approx. 70000 finished processes).In interactive mode a warning is given whenever processes have been skippedfor this reason.

 

COLORS

For the resource consumption on system level,atopuses colors to indicate that a critical occupation percentage hasbeen (almost) reached. A critical occupation percentage means that is likely that this loadcauses a noticeable negative performance influence for applications usingthis resource. The critical percentage depends on the type of resource:e.g. the performance influence of a disk with a busy percentage of 80%might be more noticeable for applications/user than a CPU with a busypercentage of 90%.
Currentlyatopuses the following default values to calculate a weighted percentageper resource:

 Processor
A busy percentage of 90% or higher is considered `critical'.
 Disk
A busy percentage of 70% or higher is considered `critical'.
 Network
A busy percentage of 90% or higher for the load of an interface isconsidered `critical'.
 Memory
An occupation percentage of 90% is considered `critical'.Notice that this occupation percentage is the accumulated memoryconsumption of the kernel (including slab) and all processes; thememory for the page cache (`cache' and `buff' in the MEM-line) and thereclaimable part of the slab (`slrec`) is not implied!
If the number of pages swapped out (`swout' in the PAG-line) is largerthan 10 per second, the memory resource is considered `critical'.A value of at least 1 per second is considered `almost critical'.
If the committed virtual memory exceeds the limit (`vmcom' and `vmlim'in the SWP-line), the SWP-line is colored due to overcommitting the system.
 Swap
An occupation percentage of 80% is considered `critical'because swap space might be completely exhausted in the near future;it is not critical from a performance point-of-view.

These default values can be modified in the configuration file(see separate man-page of atoprc).

When a resource exceeds its critical occupation percentage, the concerningvalues in the screen line are colored red by default.
When a resource exceeded (default) 80% of its critical percentage(so it is almost critical), the concerning values in the screen lineare colored cyan by default. This `almost critical percentage' (one valuefor all resources) can be modified in the configuration file(see separate man-page of atoprc).
The default colors red and cyan can be modified in the configuration fileas well (see separate man-page of atoprc).

With the key 'x' (or flag -x), the use of colors can be suppressed. 

NETATOP MODULE

Per-process and per-thread network activity can be measured by thenetatopkernel module. You can download this kernel module from the website(mentioned at the end of this manual page) and install it on yoursystem if the kernel version is 2.6.24 or newer.
Whenatopgathers counters for a new interval, it verifies if thenetatopmodule is currently active. If so,atopobtains the relevant network counters from this module and showsthe number of sent and received packets per process/thread in the genericscreen. Besides, detailed counters can be requested bypressing the `n' key.
When thenetatopddaemon is running as well,atopalso reads the network counters of exited processes that are loggedby this daemon (comparable with process accounting).

More information about the optionalnetatopkernel module and thenetatopddaemon can be found in the concerning man-pages and on the websitementioned at the end of this manual page. 

GPU STATISTICS GATHERING

GPU statistics can be gathered byatopgpudwhich is a separate data collection daemon process.It gathers cumulative utilization counters of every Nvidia GPUin the system, as well as utilization counters of every process that uses a GPU.Whenatopnotices that the daemon is active, it reads these GPU utilizationcounters with every interval.

Theatopgpuddaemon is written in Python, so a Python interpreter should be installedon the target system. The Python code of the daemon is compatible withPython version 2 and version 3.For the gathering of the statistics, thepynvmlmodule is used by the daemon. Be sure that this module is installedon the target system before activating the daemon, by running thecommand as rootpip(the commandpipmight be exchanged bypip3in case of Python3):

  pip install nvidia-ml-py

Theatopgpuddaemon is installed by default as part of theatoppackage, but it is notautomatically enabled.The daemon can be enabled and started now by running the following commands(as root):

  systemctl enable atopgpu
  systemctl start atopgpu

Find a description about the utilization counters in the section OUTPUT DESCRIPTION. 

INTERACTIVE COMMANDS

When runningatopinteractively (no output redirection), keys can be pressed to control theoutput. In general, lower case keys can be used to show other information forthe active processes and upper case keys can be used to influence thesort order of the active process/thread list.

g
Show generic output (default).

Per process the following fields are shown in case of a window-widthof 80 positions:process-id, cpu consumption duringthe last interval in system and user mode, the virtual and residentmemory growth of the process.

The subsequent columns depend on the used kernel:
When the kernel supports "storage accounting" (>= 2.6.20), the datatransfer for read/write on disk, the status and exit code areshown for each process.When the kernel does not support"storage accounting", the username, number of threads in thethread group, the status and exit code are shown.
When the kernel module 'netatop' is loaded, the data transfer for send/receiveof network packets is shown for each process.
The last columns contain the state, the occupation percentage for thechosen resource (default: cpu) and the process name.

When more than 80 positions are available, other information is added.

m
Show memory related output.

Per process the following fields are shown in case of a window-widthof 80 positions:process-id, minor and majormemory faults, size of virtual shared text, total virtual process size, total resident process size, virtual and resident growth duringlast interval, memory occupation percentage and process name.

When more than 80 positions are available, other information is added.

For memory consumption, always all processes are shown (also the processesthat were not active during the interval).

d
Show disk-related output.

When "storage accounting" is active in the kernel, the followingfields are shown:process-id, amount of data read from disk, amount of data written to disk,amount of data that was written but has been withdrawn again (WCANCL),disk occupation percentage and process name.

n
Show network related output.

Per process the following fields are shown in case of a window-widthof 80 positions:process-id, thread-id,total bandwidth for received packets,total bandwidth for sent packets,number of received TCP packets with the average size per packet (in bytes),number of sent TCP packets with the average size per packet (in bytes),number of received UDP packets with the average size per packet (in bytes),number of sent UDP packets with the average size per packet (in bytes),the network occupation percentage and process name.
This information can only be shown when kernel module `netatop' is installed.

When more than 80 positions are available, other information is added.

s
Show scheduling characteristics.

Per process the following fields are shown in case of a window-widthof 80 positions:process-id,number of threads in state 'running' (R),number of threads in state 'interruptible sleeping' (S),number of threads in state 'uninterruptible sleeping' (D),scheduling policy (normal timesharing, realtime round-robin, realtime fifo),nice value, priority, realtime priority, current processor,status, exit code, state, the occupation percentage for the chosenresource and the process name.

When more than 80 positions are available, other information is added.

v
Show various process characteristics.

Per process the following fields are shown in case of a window-widthof 80 positions:process-id, user name and group,start date and time, status (e.g. exit code if the process has finished),state, the occupation percentage for the chosen resource and the process name.

When more than 80 positions are available, other information is added.

c
Show the command line of the process.

Per process the following fields are shown: process-id,the occupation percentage for the chosen resource and thecommand line including arguments.

e
Show GPU utilization.

Per process at least the following fields are shown:process-id,range of GPU numbers on which the process currently runs,GPU busy percentage on all GPUs,memory busy percentage (i.e. read and write accesses on memory) on all GPUs,memory occupation at the moment of the sample,average memory occupation during the sample, andGPU percentage.

When theatopgpuddaemon does not run with root privileges, the GPU busy percentage andthe memory busy percentage are not available on process level.In that case, the GPU percentage on process level reflects theGPU memory occupation instead of the GPU busy percentage (whichis preferred).

o
Show the user-defined line of the process.

In the configuration file the keywordownproclinecan be specified with the description of a user-defined output-line.
Refer to the man-page ofatoprcfor a detailed description.

y
Show the individual threads within a process (toggle).

Single-threaded processes are still shown as one line.
For multi-threaded processes, one line represents the processwhile additional lines show the activityper individual thread (in a different color). Depending onthe option 'a' (all or active toggle), all threads are shownor only the threads that were active during the last interval.
Whether this key is active or not can be seen in the header line.

u
Show the process activity accumulated per user.

Per user the following fields are shown: number of processes activeor terminated during last interval (or in total if combined with command `a'),accumulated cpu consumption during last interval in system and user mode,the current virtual and resident memory space consumed by active processes(or all processes of the user if combined with command `a').
When "storage accounting" is active in the kernel,the accumulated read and write throughput on disk is shown.When the kernel module `netatop' has been installed,the number of received and sent network packets are shown.
The last columns contain the accumulated occupation percentage for thechosen resource (default: cpu) and the user name.

p
Show the process activity accumulated per program (i.e. process name).

Per program the following fields are shown: number of processes activeor terminated during last interval (or in total if combined with command `a'),accumulated cpu consumption during last interval in system and user mode,the current virtual and resident memory space consumed by active processes(or all processes of the user if combined with command `a').
When "storage accounting" is active in the kernel,the accumulated read and write throughput on disk is shown.When the kernel module `netatop' has been installed,the number of received and sent network packets are shown.
The last columns contain the accumulated occupation percentage for thechosen resource (default: cpu) and the program name.

j
Show the process activity accumulated per Docker container.

Per container the following fields are shown: number of processes activeor terminated during last interval (or in total if combined with command `a'),accumulated cpu consumption during last interval in system and user mode,the current virtual and resident memory space consumed by active processes(or all processes of the user if combined with command `a').
When "storage accounting" is active in the kernel,the accumulated read and write throughput on disk is shown.When the kernel module `netatop' has been installed,the number of received and sent network packets are shown.
The last columns contain the accumulated occupation percentage for thechosen resource (default: cpu) and the Docker container id (CID).

C
Sort the current list in the order of cpu consumption (default).The one-but-last column changes to ``CPU''.

E
Sort the current list in the order of GPU utilization(preferred, but only applicablewhen theatopgpuddaemon runs under root privileges) or the order ofGPU memory occupation).The one-but-last column changes to ``GPU''.

M
Sort the current list in the order of resident memory consumption.The one-but-last column changes to ``MEM''. In case of sorting on memory,the full process list will be shown (not only the active processes).

D
Sort the current list in the order of disk accesses issued.The one-but-last column changes to ``DSK''.

N
Sort the current list in the order of network bandwidth (receivedand transmitted).The one-but-last column changes to ``NET''.

A
Sort the current list automatically in the order of the most busysystem resource during this interval.The one-but-last column shows either ``ACPU'', ``AMEM'', ``ADSK'' or ``ANET''(the preceding 'A' indicates automatic sorting-order).The most busy resource is determined by comparing the weightedbusy-percentages of the system resources, as described earlier inthe section COLORS.
This option remains valid untilanother sorting-order is explicitly selected again.
A sorting-order for disk is only possible when "storage accounting" is active.A sorting-order for network is only possible when the kernel module `netatop'is loaded.

Miscellaneous interactive commands:

?
Request for help information (also the key 'h' can be pressed).

V
Request for version information (version number and date).

R
Gather and calculate the proportional set size of processes (toggle).Gathering of all values that are needed to calculate the PSIZE of a processis a relatively time-consuming task, so this key should only be active whenanalyzing the resident memory consumption of processes.

x
Suppress colors to highlight critical resources (toggle).
Whether this key is active or not can be seen in the header line.

z
The pause key can be used to freeze the current situation in order toinvestigate the output on the screen. While atopis paused, the keys described above can be pressed to show otherinformation about the current list of processes.Whenever the pause key is pressed again,atop will continue with a next sample.

i
Modify the interval timer (default: 10 seconds). If an interval timer of 0 isentered, the interval timer is switched off. In that case a new sample canonly be triggered manually by pressing the key 't'.

t
Trigger a new sample manually. This key can be pressed if the current sampleshould be finished before the timer has exceeded, or if no timer is set at all(interval timer defined as 0). In the latter caseatopcan be used as a stopwatch to measure the load being caused by aparticular application transaction, without knowing on beforehand how manyseconds this transaction will last.

When viewing the contents of a raw file, this key can be used to show thenext sample from the file.

T
When viewing the contents of a raw file, this key can be used to show theprevious sample from the file.

b
When viewing the contents of a raw file, this key can be used to branchto a certain timestamp within the file (either forward or backward).

r
Reset all counters to zero to see the system and process activity sinceboot again.

When viewing the contents of a raw file, this key can be used to rewindto the beginning of the file again.

U
Specify a search string for specific user names as a regular expression.From now on, only (active) processes will be shown from a user which matchesthe regular expression.The system statistics are still system wide.If the Enter-key is pressed without specifying a name, (active)processes of all users will be shown again.
Whether this key is active or not can be seen in the header line.

I
Specify a list with one or more PIDs to be selected.From now on, only processes will be shown with a PID which matchesone of the given list.The system statistics are still system wide.If the Enter-key is pressed without specifying a PID, all (active)processes will be shown again.
Whether this key is active or not can be seen in the header line.

P
Specify a search string for specific process names as a regular expression.From now on, only processes will be shown with a name which matches theregular expression.The system statistics are still system wide.If the Enter-key is pressed without specifying a name, all (active)processes will be shown again.
Whether this key is active or not can be seen in the header line.

/
Specify a specific command line search string as a regular expression.From now on, only processes will be shown with a command line whichmatches the regular expression.The system statistics are still system wide.If the Enter-key is pressed without specifying a string, all (active)processes will be shown again.
Whether this key is active or not can be seen in the header line.

J
Specify a Docker container id of 12 (hexadecimal) characters.From now on, only processes will be shown that run in that specificDocker container (CID).The system statistics are still system wide.If the Enter-key is pressed without specifying a container id,all (active) processes will be shown again.
Whether this key is active or not can be seen in the header line.

S
Specify search strings for specific logical volume names,specific disk names and specific network interface names. Allsearch strings are interpreted as a regular expressions.From now on, only those system resources are shown that matchthe concerning regular expression.If the Enter-key is pressed without specifying a search string, all (active)system resources of that type will be shown again.
Whether this key is active or not can be seen in the header line.

a
The `all/active' key can be used to toggle between only showing/accumulatingthe processes that were active during the last interval (default) orshowing/accumulating all processes.
Whether this key is active or not can be seen in the header line.

G
By default, atopshows/accumulates the processes that are alive and the processesthat are exited during the last interval. With this key (toggle),showing/accumulating the processes that are exited can be suppressed.
Whether this key is active or not can be seen in the header line.

f
Show a fixed (maximum) number of header lines for system resources (toggle).By default only the lines are shown about system resources (CPUs, paging,logical volumes, disks, network interfaces) that really have been activeduring the last interval.With this key you can forceatopto show lines of inactive resources as well.
Whether this key is active or not can be seen in the header line.

F
Suppress sorting of system resources (toggle).By default system resources (CPUs, logical volumes, disks,network interfaces) are sorted on utilization.
Whether this key is active or not can be seen in the header line.

1
Show relevant counters as an average per second (in the format `..../s')instead of as a total during the interval (toggle).
Whether this key is active or not can be seen in the header line.

l
Limit the number of system level lines for the counters per-cpu,the active disks and the network interfaces.By default lines are shown of all CPUs, disks and network interfaceswhich have been active during the last interval.Limiting these lines can be useful on systems with huge number CPUs,disks or interfaces in order to be able to runatopon a screen/window with e.g. only 24 lines.
For all mentioned resources the maximum number of lines can be specifiedinteractively. When using the flag-lthe maximum number of per-cpu lines is set to 0,the maximum number of disk lines to 5 andthe maximum number of interface lines to 3.These values can be modified again in interactive mode.

k
Send a signal to an active process (a.k.a. kill a process).

q
Quit the program.

PgDn
Show the next page of the process/thread list.
With the arrow-down key the list can be scrolled downwards with single lines.

^F
Show the next page of the process/thread list (forward).
With the arrow-down key the list can be scrolled downwards with single lines.

PgUp
Show the previous page of the process/thread list.
With the arrow-up key the list can be scrolled upwards with single lines.

^B
Show the previous page of the process/thread list (backward).
With the arrow-up key the list can be scrolled upwards with single lines.

^L
Redraw the screen.
 

RAW DATA STORAGE

In order to store system and process level statistics for long-termanalysis (e.g. to check the system load and the active processes runningyesterday between 3:00 and 4:00 PM),atopcan store the system and process level statistics incompressed binary format in a raw file with the flag-wfollowed by the filename.If this file already exists and is recognized as a raw data file,atopwill append new samples to the file (starting with a sample which reflectsthe activity since boot); if the file does not exist, it will be created.
All information about processes and threads is stored in the raw file.
The interval (default: 10 seconds) and number of samples (default: infinite)can be passed as last arguments. Instead of the number of samples, the flag-Scan be used to indicate thatatopshould finish anyhow before midnight.

A raw file can be read and visualized again with the flag-rfollowed by the filename. If no filename is specified, the file/var/log/atop/atop_YYYYMMDDis opened for input (whereYYYYMMDDare digits representing the current date).If a filename is specified in the format YYYYMMDD (representing any validdate), the file/var/log/atop/atop_YYYYMMDDis opened.If a filename with the symbolic nameyis specified, yesterday's daily logfile is opened(this can be repeated so 'yyyy' indicates the logfile of four days ago).
The samples from the file can be viewed interactively by using the key 't'to show the next sample, the key 'T' to show the previous sample, thekey 'b' to branch to a particular time or the key 'r' to rewind tothe begin of the file.
When output is redirected to a file or pipe,atopprints all samples in plain ASCII. The default line length is 80 charactersin that case; with the flag-Lfollowed by an alternate line length, more (or less) columns will be shown.
With the flag-b(begin time) and/or-e(end time) followed by a time argument of the form HH:MM,a certain time period within the raw file can be selected.

Whenatopis installed, the scriptatop.dailyis stored in the/usr/share/atopdirectory.This scripts takes care thatatopis activated every day at midnight to write compressed binary data to the file/var/log/atop/atop_YYYYMMDDwith an interval of 10 minutes by default. The-Rflag is passed by default to gather information about the proportionalset size of every process.
Furthermore the script removes all raw files which are by defaultolder than 28 days.
The mentioned default values can be overruled by creating the file/etc/default/atopthat might contain other values forLOGOPTS(by default the-Rflag), LOGINTERVAL(in seconds, by default 600), andLOGGENERATIONS(in days, by default 28).

Theatop.dailyscript is activated via thecrondaemon using the file/etc/cron.d/atopwith the contents
        0 0 * * * root /usr/share/atop/atop.daily

When the packagepsacctis installed, the process accounting is automatically restarted via thelogrotatemechanism. The file/etc/logrotate.d/psaccs_atoptakes care thatatopis finished just before the rotation of the process accounting fileand the file/etc/logrotate.d/psaccu_atoptakes care thatatopis restarted again after the rotation.When the packagepsacctis not installed, these logrotate-files have no effect.

Unfortunately, it is not always possible to keep the format of the raw filescompatible in newer versions ofatopespecially when lots of new counters have to be maintained.Therefore, the programatopconvertis installed to convert a raw file created by an olderversion ofatopto a raw file that can be read by a newer version ofatop(see the man page ofatopconvertfor more details).

 

OUTPUT DESCRIPTION

The first sample shows the system level activity since boot(the elapsed time in the header shows the time since boot).Note that particular counters could have reached their maximumvalue (several times) and started by zero again,so do not rely on these figures.

For every sampleatopfirst shows the lines related to system level activity. If a particular system resource has not been used during the interval, the entire linerelated to this resource is suppressed. So the number of system level linesmay vary for each sample.
After that a list is shown of processes which have been active during the lastinterval. This list is by default sorted on cpu consumption, but this ordercan be changed by the keys which are previously described.

If values have to be shown byatopwhich do not fit in the column width,another format is used. If e.g. a cpu-consumption of 233216 millisecondsshould be shown in a column width of 4 positions, it is shown as `233s'(in seconds).For large memory figures, another unit is chosen if the value does not fit(Mb instead of Kb, Gb instead of Mb, Tb instead of Gb, ...).For other values, a kind of exponent notation is used (value 123456789shown in a column of 5 positions gives 123e6). 

OUTPUT DESCRIPTION - SYSTEM LEVEL

The system level information consists of the following output lines:

PRC
Process and thread level totals.
This line contains the total cpu time consumed in system mode (`sys') and in user mode (`user'),the total number of processes present at this moment (`#proc'),the total number of threads present at this moment in state `running' (`#trun'),`sleeping interruptible' (`#tslpi') and `sleeping uninterruptible' (`#tslpu'),the number of zombie processes (`#zombie'),the number of clone system calls (`clones'), andthe number of processes that ended during the interval(`#exit') when process accounting is used. Instead of `#exit` the lastcolumn may indicate that process accounting could not be activated(`no procacct`).
If the screen-width does not allow all of these counters,only a relevant subset is shown.

CPU
CPU utilization.
At least one line is shown for the total occupation of all CPUs together.
In case of a multi-processor system, an additional line is shownfor every individual processor (with `cpu' in lower case),sorted on activity. Inactive CPUs will not be shown by default.The lines showing the per-cpu occupation contain the cpu number inthe field combined with the wait percentage.

Every line contains the percentage of cpu time spent in kernel mode by all active processes (`sys'), the percentage of cpu time consumed in user mode (`user') for allactive processes (including processes running with a nice value larger thanzero), the percentage of cpu time spent for interrupt handling (`irq')including softirq, the percentage of unused cpu time while no processeswere waiting for disk I/O (`idle'), andthe percentage of unused cpu time while at least one process was waitingfor disk I/O (`wait').
In case of per-cpu occupation, the cpu number andthe wait percentage (`w') for that cpu.The number of lines showing the per-cpu occupation can be limited.

For virtual machines, the steal-percentage (`steal') shows the percentage of cpu time stolen by other virtual machinesrunning on the same hardware.
For physical machines hosting one or more virtual machines,the guest-percentage (`guest') shows the percentage of cpu time used by the virtual machines. Notice thatthis percentage overlaps the user percentage!

When PMC performance monitoring counters are supported by the CPUand the kernel (andatopruns with root privileges), the number of instructions perCPU cycle (`ipc') is shown.The first sample always shows the value 'initial',because the counters are just activated at the moment thatatopis started.
When theCPU busy percentage is highand the IPC is less than 1.0,it is likely that the CPU is frequently waiting for memory accessduring instruction execution (larger CPU caches or faster memory mightbe helpful to improve performance).When theCPU busy percentage is highand the IPC is greater than 1.0,it is likely that the CPU is instruction-bound (more/faster coresmight be helpful to improve performance).
Furthermore, per CPU the effective number of cycles (`cycl') is shown.This value can reach the current CPU frequency if such CPU is 100% busy.When an idle CPU is halted, the number of effective cycles canbe (considerably) lower than the current frequency.
Notice that theaverageinstructions per cycle and number of cycles is shown in the CPU linefor all CPUs.
See also: http://www.brendangregg.com/blog/2017-05-09/cpu-utilization-is-wrong.html

In case of frequency scaling, all previously mentioned CPU percentagesare relative to the used scaling of the CPU during the interval.If a CPU has been active for e.g. 50% in user mode during the intervalwhile the frequency scaling of that CPU was 40%, only 20% of the fullcapacity of the CPU has been used in user mode.
In case that the kernel module `cpufreq_stats' is active(after issueing `modprobe cpufreq_stats'), theaveragefrequency (`avgf') and theaveragescaling percentage (`avgscal') is shown. Otherwise thecurrentfrequency (`curf') and thecurrentscaling percentage (`curscal') is shown at the moment that the sampleis taken.Notice thataveragevalues for frequency and scaling are shown in the CPU line for every CPU.
Frequency scaling statistics are only gathered for systemswith maximum 8 CPUs, since gathering of these values per CPUis very time consuming.

If the screen-width does not allow all of these counters,only a relevant subset is shown.

CPL
CPU load information.
This line contains the load average figures reflecting the numberof threads that are available to run on a CPU (i.e. part of the runqueue)or that are waiting for disk I/O. These figures are averaged over1 (`avg1'), 5 (`avg5') and 15 (`avg15') minutes.
Furthermore the number of context switches (`csw'), the numberof serviced interrupts (`intr') and the number of available CPUs are shown.

If the screen-width does not allow all of these counters,only a relevant subset is shown.

GPU
GPU utilization (Nvidia).
Read the section GPU STATISTICS GATHERING in this document to find the detailsabout the activation of theatopgpuddaemon.

In the first column of every line, the bus-id (last nine characters) andthe GPU number are shown.The subsequent columns show the percentage of time that one or more kernelswere executing on the GPU (`gpubusy'), the percentage of time that global(device) memory was being read or written (`membusy'), the occupationpercentage of memory (`memocc'), the total memory (`total'), the memorybeing in use at the moment of the sample (`used'), the average memorybeing in use during the sample time (`usavg'), the number of processesbeing active on the GPU at the moment of the sample (`#proc'), andthe type of GPU.

If the screen-width does not allow all of these counters,only a relevant subset is shown.
The number of lines showing the GPUs can be limited.

MEM
Memory occupation.
This line contains the total amount of physical memory(`tot'), the amount of memory which is currently free (`free'),the amount of memory in use as page cache includingthe total resident shared memory (`cache'), the amount of memory within thepage cache that has to be flushed to disk (`dirty'), the amountof memory used for filesystem meta data (`buff'), the amount ofmemory being used for kernel mallocs (`slab'), the amount ofslab memory that is reclaimable (`slrec'), the resident size of sharedmemory including tmpfs (`shmem`), the resident size of shared memory (`shrss`)the amount of shared memory that is currently swapped (`shswp`),the amount of memory that is currently claimed by vmware'sballoon driver (`vmbal`),the amount of memory that is claimed for huge pages (`hptot`),and the amount of huge page memory that is really in use (`hpuse`).

If the screen-width does not allow all of these counters,only a relevant subset is shown.

SWP
Swap occupation and overcommit info.
This line contains the total amount of swap space on disk (`tot') andthe amount of free swap space (`free').
Furthermore the committed virtual memory space (`vmcom') and the maximum limit of the committed space (`vmlim', which is by default swap sizeplus 50% of memory size) is shown.The committed space is the reserved virtual space for all allocations ofprivate memory space for processes. The kernel only verifies whether thecommitted space exceeds the limit if strict overcommit handling isconfigured (vm.overcommit_memory is 2).

PAG
Paging frequency.
This line contains the number of scanned pages (`scan') due to the factthat free memory drops below a particular threshold and the numbertimes that the kernel tries to reclaim pages due to an urgent need (`stall').
Also the number of memory pages the system read from swap space (`swin')and the number of memory pages the system wrote to swap space (`swout')are shown.

PSI
Pressure Stall Information.
This line contains three percentages per category:average pressure percentage over the last 10, 60 and 300 seconds(separated by slashes).
The categories are: CPU for 'some' (`cs'),memory for 'some' (`ms'), memory for 'full' (`mf'),I/O for 'some' (`is'), and I/O for 'full' (`if').

LVM/MDD/DSK
Logical volume/multiple device/disk utilization.
Per active unit one line is produced, sorted on unit activity.Such line shows the name (e.g. VolGroup00-lvtmp for a logical volume orsda for a hard disk), the busy percentage i.e. the portion of time that theunit was busy handling requests (`busy'), the number of read requests issued(`read'), the number of write requests issued (`write'),the number of KiBytes per read (`KiB/r'), the number of KiBytes per write (`KiB/w'), the number of MiBytes per second throughput for reads (`MBr/s'), the number of MiBytes per second throughput for writes (`MBw/s'), the average queue depth (`avq')and the average number of milliseconds needed by a request (`avio')for seek, latency and data transfer.
If the screen-width does not allow all of these counters,only a relevant subset is shown.

The number of lines showing the units can be limited per class (LVM, MDD orDSK) with the 'l' key or statically (see separate man-page of atoprc).By specifying the value 0 for a particular class, no lines will beshown any more for that class.

NFM
Network Filesystem (NFS) mount at the client side.
For each NFS-mounted filesystem, a line is shown that contains the mounted server directory, the name of the server (`srv'),the total number of bytes physically read from the server (`read') andthe total number of bytes physically written to the server (`write').Data transfer is subdivided inthe number of bytes read via normal read() system calls (`nread'),the number of bytes written via normal read() system calls (`nwrit'),the number of bytes read via direct I/O (`dread'),the number of bytes written via direct I/O (`dwrit'),the number of bytes read via memory mapped I/O pages (`mread'), andthe number of bytes written via memory mapped I/O pages (`mwrit').

NFC
Network Filesystem (NFS) client side counters.
This line contains the number of RPC calls issues by local processes (`rpc'), the number of read RPC calls (`read`) andwrite RPC calls (`rpwrite') issued to the NFS server,the number of RPC calls being retransmitted (`retxmit')and the number of authorization refreshes (`autref').

NFS
Network Filesystem (NFS) server side counters.
This line contains the number of RPC calls received from NFS clients (`rpc'),the number of read RPC calls received (`cread`),the number of write RPC calls received (`cwrit'),the number of Megabytes/second returned to read requests by clients (`MBcr/s`),the number of Megabytes/second passed in write requests by clients (`MBcw/s`),the number of network requests handled via TCP (`nettcp'), the number of network requests handled via UDP (`netudp'),the number of reply cache hits (`rchits'),the number of reply cache misses (`rcmiss') andthe number of uncached requests (`rcnoca').Furthermore some error counters indicating the number of requestswith a bad format (`badfmt') or a bad authorization (`badaut'), and acounter indicating the number of bad clients (`badcln').

NET
Network utilization (TCP/IP).
One line is shown for activity of the transport layer (TCP and UDP), one linefor the IP layer and one line per active interface.
For the transport layer,counters are shown concerning the number of received TCP segmentsincluding those received in error (`tcpi'),the number of transmitted TCP segments excludingthose containing only retransmitted octets (`tcpo'),the number of UDP datagrams received (`udpi'),the number of UDP datagrams transmitted (`udpo'),the number of active TCP opens (`tcpao'),the number of passive TCP opens (`tcppo'),the number of TCP output retransmissions (`tcprs'),the number of TCP input errors (`tcpie'),the number of TCP output resets (`tcpor'),the number of UDP no ports (`udpnp'), andthe number of UDP input errors (`udpie').
If the screen-width does not allow all of these counters,only a relevant subset is shown.
These counters are related to IPv4 and IPv6 combined.

For the IP layer, counters are shown concerning the number of IP datagramsreceived from interfaces, including those received in error (`ipi'),the number of IP datagrams that local higher-layer protocols offered fortransmission (`ipo'), the number of received IP datagrams which wereforwarded to other interfaces (`ipfrw'), the number of IP datagrams whichwere delivered to local higher-layer protocols (`deliv'),the number of received ICMP datagrams (`icmpi'), andthe number of transmitted ICMP datagrams (`icmpo').
If the screen-width does not allow all of these counters,only a relevant subset is shown.
These counters are related to IPv4 and IPv6 combined.

For every active network interface one line is shown,sorted on the interface activity.Such line shows the name of the interface and its busy percentagein the first column.The busy percentage for half duplex is determined by comparing theinterface speed with the number of bits transmitted and receivedper second; for full duplex the interface speed is compared with thehighest of either the transmitted or the received bits.When the interface speed can not be determined (e.g. for the loopbackinterface), `---' is shown instead of the percentage.
Furthermore the number of received packets (`pcki'),the number of transmitted packets (`pcko'),the line speed of the interface (`sp'),the effective amount of bits received per second (`si'),the effective amount of bits transmitted per second (`so'),the number of collisions (`coll'),the number of received multicast packets (`mlti'),the number of errors while receiving a packet (`erri'),the number of errors while transmitting a packet (`erro'),the number of received packets dropped (`drpi'), andthe number of transmitted packets dropped (`drpo').
If the screen-width does not allow all of these counters,only a relevant subset is shown.
The number of lines showing the network interfaces can be limited.

IFB
Infiniband utilization.
For every active Infiniband port one line is shown,sorted on activity.Such line shows the name of the port and its busy percentagein the first column.The busy percentage is determined by taking thehighest of either the transmitted or the received bits during the interval,multiplying that value by the number of lanes and comparing it against themaximum port speed.
Furthermore the number of received packets divided by thenumber of lanes (`pcki'),the number of transmitted packets divided by the number of lanes (`pcko'),the maximum line speed (`sp'),the effective amount of bits received per second (`si'),the effective amount of bits transmitted per second (`so'), andthe number of lanes (`lanes').
If the screen-width does not allow all of these counters,only a relevant subset is shown.
The number of lines showing the Infiniband ports can be limited.
 

OUTPUT DESCRIPTION - PROCESS LEVEL

Following the system level information, the processes are shown from which theresource utilization has changed during the last interval. These processesmight have used cpu time or issued disk or network requests. However a processis also shown if part of it has been paged out due to lack of memory (whilethe process itself was in sleep state).

Per process the following fields may be shown (in alphabetical order),depending on the current output mode as described in the sectionINTERACTIVE COMMANDS and depending on the current width of your window:

AVGRSZ
The average size of one read-action on disk.

AVGWSZ
The average size of one write-action on disk.

BANDWI
Total bandwidth for received TCP and UDP packets consumed by this process(bits-per-second).This value can be compared with the value `si'on interface level (used bandwidth per interface).
This information will only be shown when the kernel module `netatop'is loaded.

BANDWO
Total bandwidth for sent TCP and UDP packets consumed by this process(bits-per-second).This value can be compared with the value `so'on interface level (used bandwidth per interface).
This information will only be shown when the kernel module `netatop'is loaded.

CID
Container ID (Docker) of 12 hexadecimal digits, referring to the containerin which the process/thread is running.If a process has been started and finished during the lastinterval, a `?' is shown because the container ID is not part ofthe standard process accounting record.

CMD
The name of the process.This name can be surrounded by "less/greater than" signs (`<name>') which means that the process has finished during the lastinterval.
Behind the abbreviation `CMD' in the header line, the current page number andthe total number of pages of the process/thread list are shown.

COMMAND-LINE
The full command line of the process (including arguments). If the length ofthe command line exceeds the length of the screen line, the arrow keys -> and <- can be used for horizontal scroll.
Behind the verb `COMMAND-LINE' in the header line, the current page numberand the total number of pages of the process/thread list are shown.

CPU
The occupation percentage of this process related to the available capacityfor this resource on system level.

CPUNR
The identification of the CPU the (main) thread is running onor has recently been running on.

CTID
Container ID (OpenVZ).If a process has been started and finished during the lastinterval, a `?' is shown because the container ID is not part ofthe standard process accounting record.

DSK
The occupation percentage of this process related to the total load thatis produced by all processes (i.e. total disk accessesby all processes during the last interval).
This information is shown when per process "storage accounting" is activein the kernel.

EGID
Effective group-id under which this process executes.

ENDATE
Date that the process has been finished. If the process is still running,this field shows `active'.

ENTIME
Time that the process has been finished. If the process is still running,this field shows `active'.

ENVID
Virtual environment identified (OpenVZ only).

EUID
Effective user-id under which this process executes.

EXC
The exit code of a terminated process (second position of column `ST' is E)or the fatal signal number (second position of column `ST' is S or C).

FSGID
Filesystem group-id under which this process executes.

FSUID
Filesystem user-id under which this process executes.

GPU
When theatopgpuddaemon does not run with root privileges, the GPU percentagereflects the GPU memory occupation percentage (memory of all GPUs is 100%).
When theatopgpuddaemon runs with root privileges, the GPU percentagereflects the GPU busy percentage.

GPUBUSY
Busy percentage on all GPUs (one GPU is 100%).
When theatopgpuddaemon does not run with root privileges, this value is not available.

GPUNUMS
Comma-separated list of GPUs used by the process during the interval. When the comma-separated list exceedsthe width of the column, a hexadecimal value is shown.

MAJFLT
The number of page faults issued by this process that have been solvedby creating/loading the requested memory page.

MEM
The occupation percentage of this process related to the available capacityfor this resource on system level.

MEMAVG
Average memory occupation during the interval on all used GPUs.

MEMBUSY
Busy percentage of memory on all GPUs (one GPU is 100%), i.e.the time needed for read and write accesses on memory.
When theatopgpuddaemon does not run with root privileges, this value is not available.

MEMNOW
Memory occupation at the moment of the sample on all used GPUs.

MINFLT
The number of page faults issued by this process that have been solvedby reclaiming the requested memory page from the free list of pages.

NET
The occupation percentage of this process related to the total load thatis produced by all processes (i.e. consumed network bandwidthof all processes during the last interval).
This information will only be shown when kernel module `netatop' is loaded.

NICE
The more or less static priority that can be given to a process on ascale from -20 (high priority) to +19 (low priority).

NPROCS
The number of active and terminated processes accumulated for this useror program.

PID
Process-id.If a process has been started and finished during the lastinterval, a `?' is shown because the process-id is not part ofthe standard process accounting record.

POLI
The policies 'norm' (normal, which is SCHED_OTHER), 'btch' (batch)and 'idle' refer to timesharing processes.The policies 'fifo' (SCHED_FIFO) and 'rr' (round robin, which is SCHED_RR)refer to realtime processes.

PPID
Parent process-id.If a process has been started and finished during the lastinterval, value 0 is shown because the parent process-id is not part ofthe standard process accounting record.

PRI
The process' priority ranges from 0 (highest priority) to 139 (lowestpriority). Priority 0 to 99 are used for realtime processes (fixedpriority independent of their behavior) and priority 100 to 139 fortimesharing processes (variable priority depending on their recentCPU consumption and the nice value).

PSIZE
The proportional memory size of this process (or user).
Every process shares resident memory with other processes. E.g. when aparticular program is started several times, the code pages (text) areonly loaded once in memory and shared by all incarnations. Also the codeof shared libraries is shared by all processes using that shared library,as well as shared memory and memory-mapped files.For the PSIZE calculation of a process, the resident memory of a processthat is shared with other processes is divided by the number of sharers.This means, that every process is accounted for a proportional part ofthat memory. Accumulating the PSIZE values of all processes in thesystem gives a reliable impression of the total resident memory consumedby all processes.
Since gathering of all values that are needed to calculate the PSIZE is arelatively time-consuming task, the 'R' key (or '-R' flag) shouldbe active. Gathering these values also requires superuser privileges(otherwise '?K' is shown in the output).
If a process has finished during the last interval, no value is shownsince the proportional memory size is not part of the standardprocess accounting record.

RDDSK
When the kernel maintains standard io statistics (>= 2.6.20):
The read data transfer issued physically on disk (so reading from thedisk cache is not accounted for).
Unfortunately, the kernel aggregates thedata tranfer of a process to the data transfer of its parent process whenterminating, so you might see transfers for (parent) processes likecron, bash or init, that are not really issued by them.

RGID
The real group-id under which the process executes.

RGROW
The amount of resident memory that the process has grown during the lastinterval. A resident growth can be caused by touching memory pages whichwere not physically created/loaded before (load-on-demand). Note that a resident growth can also be negative e.g. when part of the processis paged out due to lack of memory or when the process frees dynamically allocated memory.For a process which started during the last interval, the resident growthreflects the total resident size of the process at that moment.
If a process has finished during the last interval, no value is shownsince resident memory occupation is not part of the standardprocess accounting record.

RNET
The number of TCP- and UDP packets received by this process.This information will only be shown when kernel module `netatop' is installed.
If a process has finished during the last interval, no value is shownsince network counters are not part of the standard process accounting record.

RSIZE
The total resident memory usage consumed by this process (or user).Notice that the RSIZE of a process includes all resident memory usedby that process, even if certain memory parts are shared with other processes(see also the explanation of PSIZE).
If a process has finished during the last interval, no value is shownsince resident memory occupation is not part of the standardprocess accounting record.

RTPR
Realtime priority according the POSIX standard.Value can be 0 for a timesharing process (policy 'norm', 'btch' or 'idle')or ranges from 1 (lowest) till 99 (highest) for a realtime process(policy 'rr' or 'fifo').

RUID
The real user-id under which the process executes.

S
The current state of the (main) thread: `R' for running(currently processing or in the runqueue), `S' for sleeping interruptible(wait for an event to occur), `D' for sleeping non-interruptible, `Z' for zombie (waiting to be synchronizedwith its parent process), `T' for stopped (suspended or traced), `W' forswapping, and `E' (exit) for processes which have finished during the lastinterval.

SGID
The saved group-id of the process.

SNET
The number of TCP and UDP packets transmitted by this process.This information will only be shown when the kernel module `netatop'is loaded.

ST
The status of a process.
The first position indicates if the process has beenstarted during the last interval (the value Nmeans 'new process').

The second position indicates if the process has beenfinished during the last interval.
The valueEmeans 'exit' on the process' own initi